Horizontonal artifacts during Windows logo

[ultron]

I recently bought ASUS R9 380 STRIX and i installed it couple days ago. First time i seated the card and booted the PC i saw horizontonal red lines on the screen, then the "no signal" message appeared the screen. I shut down PC and reseated DVI cable, then the problem has been solved. I tested the card on stock speed and i didn't see any problems. Then i tried ASUS GPU Tweak's OC mode yesterday. I get a BSOD on GTA V. Today i booted up PC and on Windows logo a horizontonal line appeared but it wasn't look like the first artifacts. I can't describe it easily but if i see it again i'll shoot a video about it. My questions:

1/Does it relate the overclocking? I did it with ASUS's software and it was very small overclock. And if overclocking was unsuccessful why i didn't have problems on stress tests?

2/Or my cable is bad, or there's a problem with the connector. I'll try it with a brand new HDMI cable.

I have read this useful article http://www.playtool.com/pages/artifacts/artifacts.html but i can't match any of this artifacts.

 

[Adored]

Make sure the card is firmly seated in the slot but yes disable the overclock and see if it stops. You might need to give it some more voltage but my overall feeling is "bad card".

 

[Ed1]

It wouldn't be GPU tweak at bootup , bios/logo screen cause the app didn't load yet till you get into windows IMO.

 

[ultron]

I didn't shot a video about it yet, but can refresh rate/resolution change cause this?

 

[Adored]

Yes you're right, it won't be causing the windows screen errors. It could be affecting GTAV but overall it looks more like a bad card or some other hardware issue.

 

[Magic Carpet]

Without further ado, looks like a bad card to me. My recently failed card (GTX 670) had similar symptoms. GTA V killed it, I say!

I hope, your "failing" card is still under the warranty, because this is exactly when one uses it!

 

[ultron]

I turned back to Gaming Mode and now it happened again. Same time and identically same visual corruption. Now I'm sure that the bad cable is not a possibility. I shot a video(sorry about the quality it captured with a cellphone camera but the artifact can seen easily):

http://vimeo.com/162369973#t=0m29s

Something looks interesting to me:

*the vertical line is blue and i'll watch it very carefully if it's gonna happen next time but it looks like same color tone next choose operating system screen. Coincidence? Does something slows the screen passes and a tearing happens? The countdown must be 3 seconds but this screen fades faster. If it happens in previous time and previous card, this can be high possibility. But i don't installed a software which can affect startup except GPU Tweak.

 

[ultron]

Without further ado, looks like a bad card to me. My recently failed card (GTX 670) had similar symptoms. GTA V killed it, I say!

I hope, your "failing" card is still under the warranty, because this is exactly when one uses it!

Most of the people claim that GTA V is a great stress test for GPUs. They overclock their cards, they run stress tests long hours and their GPUs pass the test successfully. But when they play this game; BSODs, crashes, black screens and artifacts happen which they are not existent on stock speeds.

Yes, my card is under warranty but actually i don't want to waste my time dealing with it.

By the way, i am very curious about why the most of the graphics card deaths happen on gaming/under high load? Some cards are never overheats, runs on stock speeds and they operate very cool, they run with high quality PSUs but they die on gaming. I don't see any statistics about it, but i read many forum posts about users' graphic card fails and that's my observation. Do manufacturers push their products very hard and take a small risk?

 

[Elixer]

That is either corrupted system files/drivers, or bad card.

Normally, the OEMs have a "special" version of OC'ers that have more cooling / better components and the like.
While you may get away with OC'ing ones that aren't tailored for OC'ing, you tend to see random corruption going on if you push the card too hard.

 

[ultron]

I reinstalled Windows and i never get the same artifact again. I was glad to see this but today i get two crashes. Two of them occured after i quit the game. 1st i played Black Ops 3 and after one minute i exit, screen went black. After restart i saw the driver crash report on event viewer. Then i try Arkham Knight. I played the game without crash but the same pattern revealed: After i quit this game, screen says "no signal", i resetted PC and event viewer reports "display driver stopped responding" crash. I think this is strange. I decided to deal with RMA if i can't solve the problem. I'm very curious, why the game doesn't crash, but quitting game causes this type of crashes. (Ridicously same time, not early nor not late) Card can faulty, drivers can faulty but why these problems happens only specific conditions?

 

[Elixer]

If you want to see if the system is at fault, you need to start with the basics, and test RAM with memtest86+ (let it run all tests overnight), then on 0 errors, use OCCT for a couple hours to see if CPU is OK, and you can also monitor voltages as well with that, then if that is OK, then you run furmarks and see if the video card acts up then...
If all those pass, check SMART status of your storage device.

 

[Ferzerp]

All this over that one little line that shows up during the monitor mode change during boot up and nowhere else?

 

[ultron]

I reinstalled OS and start up artifact disappeared. Now i have a new and very strange thing:

Some games crashing with "display driver stopped responding..." message. And if i right click on desktop and click "AMD Radeon Settings" i get black screen and i have to reset PC. There's strange pattern here:

*Yesterday i played Batman Arkham Knight. I get "display driver..." message and then one minute after i clicked Radeon Settings. The screen went black.

*After another AK crash(I had multiple crashes) i clicked here again and the screen went black again. There were no way but reset the PC.

*I realized that it's not a coincidence. I played Arkham Knight again, wait to crash. The game crashed with same error. Then i tried the same thing. Guess what happened: Screen goes black!

Today The Witness crashed with same error and i tried to recreate same problem. Clicked Crimson Settings and same thing happened!

I am very confused. Everyone is saying the same advices but there's strange things here. If i RMA the card and then if they tell me it's physically OK i'll waste my time only. I ran Furmark by hours and nothing is happening but if i play a graphically intensive game, it crashes with same error.

None of the stress tests shows any error.

 

[nerp]

Just RMA the card. Include screenshots of the errors if you can.

 

[ultron]

Just RMA the card. Include screenshots of the errors if you can.

The problem is they will apply classic test methods and if they couldn't detect any failure they'll give me the same card again.

I found this by the way:

https://www.reddit.com/r/AMDHelp/comments/42vmg8/the_witness_crashes_minutes_after_launch/